NAD+ Replenishment Improves Lifespan and Healthspan in Ataxia Telangiectasia Models via Mitophagy and DNA Repair

Abstract

Ataxia telangiectasia (A-T) is a rare autosomal recessive disease characterized by progressive neurodegeneration and cerebellar ataxia. A-T is causally linked to defects in ATM, a master regulator of the response to and repair of DNA double-strand breaks. The molecular basis of cerebellar atrophy and neurodegeneration in A-T patients is unclear. Here we report and examine the significance of increased PARylation, low NAD+, and mitochondrial dysfunction in ATM-deficient neurons, mice, and worms. Treatments that replenish intracellular NAD+ reduce the severity of A-T neuropathology, normalize neuromuscular function, delay memory loss, and extend lifespan in both animal models. Mechanistically, treatments that increase intracellular NAD+ also stimulate neuronal DNA repair and improve mitochondrial quality via mitophagy. This work links two major theories on aging, DNA damage accumulation, and mitochondrial dysfunction through nuclear DNA damage-induced nuclear-mitochondrial signaling, and demonstrates that they are important pathophysiological determinants in premature aging of A-T, pointing to therapeutic interventions.

Conflict of interest statement

Competing interests statement

The Bohr laboratory has CRADA arrangements with ChromaDex and GlaxoSmithKline. DAS is a consultant for GSK, Ovascience and Metrobiotech.

Published by Elsevier Inc.

Figures

Figure 1. Compromised NAD+/SIRT1 signaling induces MT dysfunction and impairs neuronal development in ATM-deficient primary neurons

(A-C) Relative MT membrane potential (A), MT ROS (B), and MT content in control and ATM-KD rat cortical neurons with different treatments. Values are the mean ± S.E.M. (n = 6-10 cultures/group). (D) Immunoblot of protein levels from control and ATM-KD neurons after various treatments. To detect acetylated PGC-1a, PGC-1a was immunoprecipitated then blotted for acetyl lysine. (E) Relative cellular NAD+ levels in control and ATM-KD rat cortical neurons with different treatments. Values are the mean ± S.E.M. (n = 6-10 cultures/group). (F) Effects of NAD+ supplementation with NR on etoposide (Et)-induced apoptotic cell death, with cleaved caspase-3 staining, in control and ATM-KD neurons. (G) Representative confocal microscopy images of control and ATM-KD neurons stained for the neuronal marker TUJ-1 and the DNA double-strand break (DSB) marker 53BP1. (H) Levels of the indicated proteins were evaluated by immunoblot analysis in lysates of cultured control and ATM-KD neurons. Veh, vehicle (DMSO); NR, nicotinamide riboside, 500 μM; NMN, nicotinamide mononucleotide, 500 μM; Resv, resveratrol, 5 μM; SRT1720/SRT, 2 μM; Ola, Olaparib, 500 nM; Veli, Veliparib, 500 nM. *p < 0.05, **p < 0.01,***p < 0.001, and n.s., not significant. See also Figure S1.

Figure 2. Restoration of the NAD+/SIR-2.1(SIRT1) signaling extends both lifespan and healthspan in short-lived atm-1 worms

(A-C) Lifespan of N2 (wild type) or atm-1 worms treated with NR, SRT, Ola, or NR+SRT beginning at the L4 stage (n = 150 worms). (D) Swimming movement rates of D10 adult worms of the indicated strains after different treatments (mean ± S.E.M, n = 15-20 worms/group). (E-F) Long-term associative memory (LTAM) and short-term associative memory (STAM) in adult D4 N2 and atm-1 worms in the different treatment groups (mean ± S.D., n = 60 worms). (G-H) Effects of NR, SRT, and Ola on muscle MT network morphology of adult D4 N2 and atm-1 worms. A myo-3∷gfp reporter gene was expressed in both nucleus and mitochondria to mark non-pharyngeal body wall muscle cells. Arbitrary score of MT network (mean ± S.E.M.; n = 53-67 muscle cells from 20 worms) (G) and representative confocal images of each condition are presented (H). MT morphology was scored on a scale from 1-5 where the value 1 denotes a severely impaired MT network, and 5 indicates a very well-organized MT network. (I) Representative thin-section electron micrographs of D7 N2 and atm-1 worms with different treatments. Cu, cuticle; Hy, hypodermis; M, muscle sarcomeres; Mito, mitochondrion. (J) Representative immunoblots from young (D1), middle aged (D10) and old (D17) N2 and atm-1 worms. For all the experiments, worms from L4 stage were exposed to vehicle control, NR (500 μM), SRT (10 μM), or Ola (500 nM). For panels D-G, *p < 0.05, **p < 0.01, or ***p < 0.001. See also Figure S2.

Figure 3. Improved NAD+/SIR-2.1(SIRT1) signaling rescues the transcriptomic and proteomic phenotypes of atm-1 nematodes

(A) Diagram showing unique and shared genes in N2 and atm-1 worms. For numbers, red denotes upregulated genes while blue denotes downregulated genes. Numbers on the line between the linked circles show shared numbers of genes. Acronyms for all 16 groups of treatments are denoted. (B-D) There are 49 common pathways induced by all the three compounds (NR, SRT and Ola). For each pathway, at least one of the compounds induced a change of Z-score ≥ |±1.5|. Venn diagram analysis of these pathways shows 16 shared pathways with 2 up-regulated and 14 down-regulated. (E) The three protein selection criteria for NR/SRT/Ola-treated D10 atm-1 worms are shown along with a list of proteins which fulfilled these criteria. Each protein was ascribed a molecular function and are further described in Table S2. (F) Venn diagram of proteomic results showing that aging has a more profound effect on atm-1 than N2 worms. (G) Venn diagram showing that NR induces more protein changes in atm-1 worms compared to N2 worms. (H) Biochemical process analysis of proteome-wide results from NR-treated D10 atm-1 worms reveals multiple health benefits after augmentation of the NAD+/SIR-2.1 pathway. See also Figure S2; Tables S1-4.

Figure 4. Reduction of the NAD+/SIRT1 pathway inhibits mitophagy and perturbs MT homeostasis in ATM-deficient neurons and worms

(A) Electron microscopy images showing MT morphology of control transfected and ATM-KD SH-SY5Ycells after treatment with vehicle or 500 μM NR. Mitochondria likely undergoing mitophagy (yellow arrow) and damaged mitochondria (red arrow) were marked. (B) Quantification of MT length from images such as those shown in panel A. Data are the mean ± S.D. (306 - 514 mitochondria in cells from 3 separate cultures). (C-D) Detection of mitophagy using a mt-mKeima fluorescent reporter in HeLa cells with/without NR treatment (500 μM). FCCP (30 μM for 3 h) was as positive control. Cells were imaged by confocal microscopy (C) and quantified for mitophagy (D). (E) A proposed working model for the mechanism of how the NAD+/SIR-2.1 pathway promotes MT maintenance. (F) Effects of NR (500 μM) on the induction of mitophagy in N2 and atm-1 worms. A MT toxicant paraquat (1 mM) was used as a positive control. Data are the mean ± S.E.M. (n=37- 57 muscle cells/group from 20 worms). (G) Relative DCT-1 protein levels were determined by quantifying levels of DCT-1 associated fluorescence (mean ± S.D., n = 10). (H) Relative mRNA levels of dct-1. Data are the mean ± S.D. from three groups of worms. (I) Contribution of NR-induced mitophagy to the improvement of healthspan (pharyngeal pumping) in N2 and atm-1. Vehicle- or NR-supplemented worms were fed with control RNAi or RNAi targeting the designated mitophagy genes at the egg hatching stage, followed by analysis of pharyngeal pumping on D7 adult worms. *p < 0.05, **P < 0.01, or ***p < 0.001, and n.s., not significant. See also Figures S3-4.

Figure 5. Restoration of the NAD+/SIRT1(SIR-2.1) pathway improves DNA repair in both ATM-deficient primary neurons and nematodes

(A) Representative immunoblots from control and ATM-KD SH-SY5Y cells showing the effects of enhancement of the NAD/SIRT1 pathway on deacetylation of Ku-70 and autophosphorylation of DNA-PKcs. (B) Representative immunoblots from control and ATM-KD SH-SY5Y cells showing that NAD+ increases the accumulation of Ku70 and DNA-PKcs on chromatin. Soluble nuclear (S.N.) fractions and chromatin fractions were isolated by a commercial kit followed by detection of protein levels by immunoblot. (C) NHEJ DNA repair efficiency was measured in neurons that were transfected with a predigested NHEJ reporter construct 3 days after either scrambled siRNA, or siRNAs targeting the designated genes. Over 100 neurons/group were counted and data shown are mean ± S.E.M. (D-E) Effects of restoration of the NAD+/SIR-2.1 pathway on genomic instability revealed by brood size (D) and male frequency (E). The atm-1(gk186) worms were used for brood size counting (D), while late generation atm-1(gk186);(h2681) worms were used for male frequency (E). Data are mean ± S.E.M. (n = 20-30 worms/group, three replicates). (F) Embryonic NHEJ capacity was measured by scoring the percentage of late stage embryos that could reach to the L4 stage 48 h after irradiation with 90 Gy. Two homologous recombination mutants, brc-1(tm1145) and brd-1(dw1) were used as negative controls, while two NHEJ mutants, cku-70(tm1524) and cku-80(ok861), were positive controls. About 150~300 embryos/group were scored, and data shown are mean ± S.E.M.

Figure 6. NAD+ improves metabolic profiles in atm-1 worms and Atm−/− mice

(A) Proteomics data showing changes of metabolism-related proteins which fulfilled the ‘3C’ standards. For ‘3C’ standards, see Fig. 3E. Each protein was ascribed a molecular function and are further described in Table S2. (B) Unsupervised hierarchical clustering of Day 6 worm metabolites among different groups. Whole worm tissues were subjected to ALEX-CIS GCTOF mass spectrometry with more than 156 metabolites identified (Table S5). (C) Changes of TCA cycle and catabolism of amino acids in atm-1 (veh) compared with N2 (veh). Amino acids in green are ketogenic, in red are glucogenic, and dark ones are involved in both glucogenic and ketogenic processes. (D-G) Changes of levels of designated metabolites in N2 and atm-1 with/without NR/NMN treatment. (H-J) Proteomics data showing changes of designated proteins. (K) Metabolome profiles of mouse cerebellum tissues performed using capillary electrophoresis/time-of-flight mass spectrometry (CE-TOFMA) and capillary electrophoresis- triple quadrupole mass spectrometry (CE-QqQMS) analysis, followed by principle component analysis (n = 3 mice/group). See also Tables S5.

Figure 7. Increasing NAD+ levels extends lifespan and ameliorates A-T phenotypes through SIRT1 activation in Atm−/− mice

(A-C) Effects of 14 days of NR or NMN supplementation on the NAD+ levels (A), rotarod performance (B), and spontaneous alternation performance (SAP) in the Y maze (C). Data shown are mean ± S.E.M. (n = 3-5 mice/group). (D) Representative confocal microscopy images of the Purkinje cells in the cerebellum of Atm−/− mice and their WT littermates after 14 days of NR or NMN treatment. Over 10 cerebellar areas for each mouse were analyzed (3-5 mice/group). Quantification of images is shown in Fig. S5A. (E) Immunoblot analysis of the indicated proteins from the cerebellum of Atm−/− mice and their WT littermates after 14 days of treatment with veh, NR or NMN. Quantification of data is in Fig. S6. (F) Representative electron microscopy images showing mitochondria in cerebellum tissues of Atm−/−mice and their WT littermates after 14-day veh, NR or NMN treatment (n = 3-5 mice/group). Damaged mitochondria were marked with red arrows. Quantification of data is in Fig. S5B-C. (G) Kaplan-Meier survival curves of vehicle and NR-treated mice. One month-old Atm−/− mice and their age- and sex-matched WT littermates were exposed to NR (12 mM in drinking water; approximately 570-590 mg/kg/day/mouse) and lifespan was determined (12-20 mice/group). Statistical analysis was performed by the log-rank test. *p < 0.05, **p < 0.01, or ***p < 0.001. See also Figures S5-S6.